Radiation dose distribution in human kidneys by octreotides in peptide receptor radionuclide therapy.

UNLABELLED Ex vivo autoradiographs of healthy kidney tissue from patients who received (111)In-DTPA-octreotide (DTPA is diethylenetriaminepentaacetic acid) before nephrectomy showed very heterogeneous radioactivity patterns in the kidneys. The consequences of the reported inhomogeneities have been evaluated for radionuclide therapy with (90)Y- DOTA-Tyr(3)-octreotide (DOTA is 1,4,7,10-tetraazacyclododecane-N,N',N'',N'''-tetraacetic acid), (177)Lu-DOTA-Tyr(3)-octreotate, and (111)In-DTPA-octreotide by calculating dose distributions and dose-volume histograms (DVHs) for the kidneys. METHODS Monte Carlo radiation transport calculations were performed by using the MCNP code version 5. The autoradiography data were used in a 2-dimensional model of the kidney tissue sections. A voxel structure inside the MIRD Pamphlet 19 multiregion kidney model was developed to generate a 3-dimensional representation of the autoradiographs. Dose distributions were calculated for the beta-emitter (90)Y, the low-energy electron and gamma-emitter (111)In, and the beta- and gamma-emitter (177)Lu. Isodose curves were generated for the 2-dimensional kidney sections and DVHs for the 3-dimensional kidney model. RESULTS The isodose curves for the high-energy beta-emitter (90)Y did not show a sign of the inhomogeneous activity distribution, apart from the cortex-medulla boundaries. Both (111)In and (177)Lu isodose curves follow the autoradiographic activity distribution exactly. The 2 gamma-rays from (111)In give higher doses to the low-radioactivity regions in the kidney sections. The DVHs show that the inhomogeneous activity distribution creates considerable volumes within the kidney and within the cortex with lower doses than the average kidney dose, together with volumes receiving much higher doses. This effect is most profound for (177)Lu, but also (111)In shows this heterogeneity in the dose distribution. CONCLUSION Kidney dosimetry for radionuclide therapy can be based on average MIRD-based dose models for high-energy beta-emitters (such as (90)Y). In contrast, low-energy beta-emitters (such as (177)Lu) and Auger-electron-emitting radionuclides (such as (111)In) produce dose distributions in the kidneys that are very dependent on the activity distribution pattern in the kidney or renal cortex. Complication probability models for renal tissue damage after radionuclide therapy with these latter nuclides need to be developed, as the existing models based on average dose to the kidney or cortex from external beam therapy experience are most probably not valid.